Favoritism,Bias, and Error in Performance Ratings of Scientists and Engineers: The Effects of Power,Status, and Numbers

In this paper we argue that the sociostructural position of groups must be taken into consideration along with motivational and cognitive processes to explain evaluations received and made by women, Blacks, Hispanics, and Asians. With this framework, we analyze performance ratings for a sample of 2,445 scientists and engineers from 24 U.S. companies and find that (a) there is more evidence of in-group favoritism than of out-group derogation; (b) high status, dominant, and majority group members enjoy favoritism expressed as a global prototype of them as competent; and (c) subordinate, minority group members overshoot in opposite ways toward other groups depending on their status and the status level of the target group. We find these effects even after controlling for self-reported productivity and for various errors inherent in the evaluation process.